Who attended the EDIS Symposium?

EDIS Symposium 2019 Blog – Part 2

Welcome to the second of a series of blogs we will be producing about the 2019 EDIS Symposium. We know that attending and presenting at research conferences is vital to career progression, personal and professional development. We also know that within the science and health research sector, some people are systematically excluded from attending, contributing at or participating fully in these events. We want to explain some of the ideas, decisions and methods we used to try to make our event inclusive, and to measure if we were successful. We asked attendees to fill out an anonymous diversity monitoring form at the EDIS symposium 2019 and here we share our findings.

In the first part of this blog series we spoke about how to collect diversity monitoring data. We included the useful Diversity and Inclusion Survey (DAISY) question guidance which was developed by the Wellcome Trust Diversity & Inclusion team, and trialled and reviewed at our 2019 Symposium. We think transparency is important, so in this blog we’re sharing our aggregated data from the EDIS Symposium 2019 and what we learnt.

We asked attendees to fill out an anonymous diversity monitoring form on the day and have collated these responses. These forms were separate to our feedback forms, so we can’t compare the experiences of different groups. To show the difference missing data can make, we’ve included the percentage of ‘non-respondents’ in the first graph about the age of delegates. These were people who attended the symposium but did not hand back their diversity monitoring form. We can’t know who these people are or how they would have identified. Filling in the diversity monitoring form was self-selective so there could be some bias as to who completed it and who didn’t, but again we can’t know this for sure.

We had a 48% response rate.Increasing this response rate will be key to the reliability and accuracy of our data, and we will be looking at strategies to do this at future events.

Please note that these questions were taken from a previous version of the DAISY guidance and therefore the updated document found at the bottom of our last blog will be slightly different. We recommend that you use this updated version of the guidance.

Quick tip: we’ve visualised our data in graphs in ways to help everyone understand the data. We’ve used both colours and patterns to indicate discrete categories depending on the graph type. This is useful for people who can’t see differences between colours and in general when printing without colour. We’ve also described the data in the alternative text of every graph.

We asked about the following demographic categories as it’s important to make sure that the topic, ethos and organisation of the EDIS symposium did not prevent individuals from any group from attending. We asked about characteristics protected by law (UK Equality Act 2010) and additional questions on caring responsibilities and Socio-Economic Status (SES). We will continue to ask these questions at all our events so that we can internally benchmark them.

1. Age

Why did we ask about age?

We were interested to know whether the age of attendees at the symposium was comparable to that of our target audience (biomedical, medical, preclinical, clinical and health researchers, as well as funders and policy makers).

What did we learn?

The age bracket with the highest representation at the EDIS symposium was up to 30 years, with a decrease in representation as age increases. We saw some representation of people above the UK retirement age (65).

Differences in the age of our attendees can come from how we reach our audiences, the topics or our approach. To check if this spread was representative of our target audience, we looked for comparable data sets for benchmarking.

One way we tried to benchmark this data was against the HESA staff and student data for the Biosciences subject area. Using both the staff and student data in the reports published by AdvanceHE, we can get a snapshot of the Biosciences’ “academic staff” and “research postgraduate students” by age group, a key community we were hoping to reach. However, the age bracket categories for the staff and student data are incomparable; the staff data uses age groupings of 10 years from 30 years old up, whereas the student data groups into varying size categories of 21 and under, 22-25, 26-35, and 36 and over. We’ve therefore presented the combined staff and student data by equally splitting student data from their original categories into their possible age groupings as used in the staff data. This is not perfect and therefore only a rough comparator. This is a reminder of why consistency in how we ask diversity monitoring questions is so important.

2. Disability

Why did we ask about disability?

We asked a series of 3 different questions relating to disabilities and long-term health conditions. We wanted to understand whether we had made our symposium accessible to people with disabilities and long-term health conditions, and we wanted to understand the differences in reporting rates for questions in this category.

What did we learn?

Only 10% of respondents considered themselves to be a disabled person, whereas when we asked if people met the Equality Act’s definition of a disability, this increased to 14%. There are many possible reasons for this – for example, some people reject the language around disability entirely. However, for our third question (where we listed possible disabilities, long-term health conditions, mental health conditions or impairments), 45% of respondents said that they had at least one of the listed conditions.

According to the Department for Work and Pensions, an estimated 19% of working-age adults in the UK have a disability under the Equality Act 2010 definition. According to the HESA data for 2017/18, 3.1% of Biosciences staff were disabled.

We know that non-disclosure rates for questions around disabilities are normally high. We were pleased to see that there wasn’t a significant increase in the ‘prefer not to say’ responses in this question compared to the rest of the survey, and we hope that the three questions together allowed people to feel comfortable disclosing any of the listed conditions as well as having the opportunity to self-define.

3. Ethnicity

Why did we ask about ethnicity?

We already know that there is under-representation in science and health research in the UK of ethnicities other than White and wanted to see if the EDIS symposium attendees reflected this.

What did we learn?

75% of EDIS symposium attendees identified their ethnic group as White. According to the UK census data for England and Wales (2011), 86% of the population is White.

In the UK HESA data for Biosciences (2017/18 academic year), only 9% of Academic Staff were BAME, though this rises slightly to 11% when looking at all SET (Science, Engineering and Technology) subjects together. For the same year, 14% of research postgraduate students in the Biological sciences in the UK were BAME and this rises to 18% for all SET subjects.

Respondents to the diversity monitoring survey at the EDIS symposium are a more ethnically diverse group than the benchmarking data we have. However, with only a 48% response rate to the diversity monitoring data, we’re limited in our ability to draw conclusions from this data.

4. Gender

Why did we ask about gender?

We had a hypothesis before the symposium that more women than men were interested and actively participating in equality, diversity and inclusion (EDI) work in the sector. We have anecdotally heard of and been a part of events about EDI where there is a much higher proportion of women present than any other gender. This is in contrast to the gender split in other scientific conferences and meetings. For example, EDIS member Connecting Science at Wellcome Genome Campus (GRL) saw a ~50/50 men to women split (data unavailable for other gender identities) in attendees across all of their 2018-19 meetings (all meetings were in the range 43-57%). We wanted to see if this was the case at the EDIS symposium, especially as we hoped to engage a more gender representative audience by focusing on the research content aspects of the topics.

What did we learn?

86% of attendees described themselves as women. When we look at conferences and events with a purely scientific content focus we see that we had a large gender bias towards women in our attendees. The Royal Society’s data from 2018 for prize lectures (48% women), scientific meetings (41% women), and public events (47% women) show women in the minority. It should be noted that the RS asks the question ‘what is your gender?’ and gives the options as ‘Male/Female/Other gender/Prefer not to say’ despite Male and Female being sex categories and not gender identities. In addition, there is no ability to self-describe. We see this approach on multiple diversity monitoring questionnaires across the sector and do not recommend it, as it is an incorrect way of conflating sex and gender identity.

We did not have any respondents identify as trans, however we want to make sure that our events and topics are trans inclusive. This is one of the reasons we addressed trans and non-binary identities specifically in our event-specific inclusion statement. We don’t know if trans people didn’t feel comfortable attending the symposium or disclosing their identity, so we will continue to monitor this data and work with the trans community to make sure our events are inclusive.

5. Sexual Orientation

Why did we ask about sexual orientation?

It’s important to make sure that the topic and ethos of the EDIS symposium did not prevent LGBQA+ individuals from attending or feeling included.

What did we learn?

Our data shows that 19% of respondents identified as Lesbian, Gay, Bisexual, Queer, Asexual or other non-heterosexual identities (LGBQA+). This is higher than official census data from the Office for National Statistics (2.7% LGBQA+ and 4.1% do not know or prefer not to say in the UK in 2017). Stonewall and the government suggest and agree that 5-7% is a reasonable estimate.  

Even if all of the people who did not return their diversity monitoring forms (48%) selected heterosexual, we still would have 9% LGBQA+ attendees as a minimum. Therefore, we think that we have either: created an event where more LGBQA+ attendees are interested in the topics, created an event where more LGBQA+ people feel comfortable attending, created an event where more LGBQA+ people feel comfortable being open about their sexual orientation on diversity monitoring forms, or a mix of any of these three reasons.

6. Caring responsibilities

Why did we ask about caring responsibilities?

No attendees requested financial support to cover caring responsibilities prior to the symposium. We wanted to know if this meant that there were no carers attending the symposium.

What did we learn?

At the EDIS symposium, 23% of respondents had caring responsibilities. Therefore, these respondents might not have needed additional financial support, might not have known that this support was available, or didn’t ask for this support.

This question currently has no comparable UK-wide benchmarking data that we can use to see if this is different from the proportion of parents and carers in the working population. We hope that other organisations use this question as described in the DAISY guidance so that we can compile a better picture of the UK scientific workforce’s caring responsibilities. We will continue to collect data related to caring responsibilities and benchmark our own events internally to make sure all parents and carers are enabled to attend.

7. Religion

Why did we ask about religion?

Religion is a protected characteristic by UK equality law. If the religious makeup of our delegates varies from what is expected, this can tell us more about the religious inclusiveness of our event.

What did we learn?

55% of attendees said they had no religion, 40% had one of 8 different religions, and 5% preferred not to say. We will continue to monitor this internally in our events and hope to work with other organisations to get a better picture of the religions and beliefs of the sector.

8. Socio-economic status (SES)

Why did we ask about SES?

Although socio-economic status isn’t a protected characteristic by UK law, there is a lot of thought currently being done to understand its impact on careers, social inclusion and discrimination. We asked two questions to look at SES based on an individual’s schooling experience and their parent’s education. Both of these are useful indicators, however there are many more questions that could be asked and we are still learning how best to measure SES. We also know that there is a gap in social mobility in the UK scientific community as only 15% of UK scientists come from a working-class household, which comprises 35% of the general population.

What did we learn?

15% of respondents attended fee paying schools whereas we know that just 7% of British people are privately educated.

The achievement gap between children of degree educated parents and those of uneducated parents has decreased but persisted in the UK for decades, with no change noted in education mobility for the least educated households. Further work needs to be done to use and combine this data across the sector to better understand the current UK scientific workforce.

What next?

Now we know who attended the EDIS 2019 symposium we can think about the different ways we advertised it as an inclusive and accessible event. Next, we’ll be exploring more of the inclusion actions we put in place and the feedback from attendees.